The Great List

The previous post was all about looking for big names in your field. Now lets talk about the next step, creating what I like to call " the state of the field" document, because you need to identify knowledge gaps and most effective ways to fill them.

Open up your favorite text editing software and write down all of the statements you can make about your field of study.

Right: X is Y when Z, or X is observed in nature often, or X and Y are observed together.

Wrong: wording your statements as a hypothesis. Your purpose here is to state what is known, not propose testable questions ( don't despair, that comes later).

Your list should be part observation and part established facts. After you do this it is time to pull together papers that substantiate each of the statements you just wrote down. Each statement should have a bunch of citations under it. Each citation should have two to three sentences that describe the major result. Here I also like to add a touch of visual organization. Simple tabs will do, just like the example bellow.

X is observed in nature.
       J.D. Haldane and P.F.Chang 1943
             - First to describe mechanism for X using infectious bacteria. Propose
constrains on X.
        S.T.Strauss et al. 2000
             - Models X as a function of Y. Equations that describe X through time are
               proposed. Documents X in ocean and terrestrial ecosystems.


Above are just a few examples, but I am trying to highlight some important points here. As you see, I chose a paper that first describes a phenomenon, the earlier the better, and a paper that shows that it is widespread in nature ( means it's important). Key is to populate this document with examples of your favorite scientific " thing" and it's description. During this process you may find that papers that you originally chose to support statement A, do a better job supporting statement B. This is great. This is the purpose of the document, to organize your thinking about the field.

Along the way you will also find that some of the statement will need rewriting. A very important thing here is honesty. You must, must, must be honest with yourself. If there is statement that you really like but can not find papers that support it, do not fall into the trap of saying "well this paper kind of supports it". "Kind of" is your worst enemy. Also do not get upset that you were wrong. If you can not find support for something, then it may be a good hypothesis to investigate. This is the second useful part of the document, it makes you see the holes.

Initially this will be the only document you need. However as your knowledge of the literature expands, and you dive further into the murky depth of scientific inquiry, this document will naturally split. Some of the original statements ( now likely refined) will form one of your projects, the rest will be better for another. Eventually you will end up with one of these documents for each project you do, and every time they will help you focus your ideas. Plus these documents make a great repository of citations for your papers.

Citation stalking -- also some stats

Fact: there is a lot of information in the world. That’s great. It is also great that my lightning fast broadband connection allows me access to it at any time of the day or night. Searching this glorious heap of facts is a different matter entirely. Google helps a great deal, even when it comes to scholarly work. However most of the time it’s just a starting point. There is something mildly uncomfortable about looking up papers that will eventually form the backbone of a thesis via Google. A solid, hard-core relational database just feels so much better.

Efficient ways of dealing with databases is something all graduate students learn eventually. There are many great databases, some of the biggest are NCBI, PubMed, EBSCO host, ISI, and many others. I am not really interested in discussing database use right now. Instead I want to discuss one very (incredibly, amazingly, fantastically, awesomely) useful feature of the ISI Web of KNowledge. Forward citation will make you happy.

Before rock-star scientists revolutionize a field, they need to know about it. Biology (or natural philosophy) has had a long and largely successful run, so “the field” is huge; navigating it can be a daunting task. Even recently invented biological pursuits like genomics, proteomics, and all of the other -omics have amassed large volumes of info. How do you find the key players, both contemporary and otherwise? To deal with this issue I like to engage in what is known (by me and my cats so far, but may become a common place name) as citation stalking.

Step one - get to a database that includes citation score with each entry. I like to use ISI Web of Knowledge because it also lets me sort by the number of citations for each publication.

Step two - pull up some papers. The more the better. This step can be a bit time consuming, ten to fifteen minutes even. You could probably watch three YouTube videos in the same amount of time, but I urge you to hold steady. Sort results by number of citations and look for a highly cited review. You want to start with a review because it will likely cite relevant articles. Remember you are not here to actually read the info, but to figure out what people to pay attention to. Every review usually covers the decade, or at least five years of info prior to its publication.

Step three - pull up citations for that review. This makes you work backwards, which is fine. You also want to work forwards as well. The greatest thing about ISI Web of Knowledge is that in addition to showing everything that a paper cites, it shows all of the papers that cite it.

Step four - rinse and repeat. Another good idea, that is after you make a list of important people in your field, is to look up their earlier work. Find out what is their claim to fame. After only a few hours of work you will be impressing your PI with mad knowledge of literature in your field, its major players and their contributions.

Since I mentioned ISI Web of Knowledge a few times, I want to leave you with the following. I typed in a few words that scientists may care about and saw what happened. I also pulled up authors of the most cited papers for each of the search-topics. Note: 10,000 is the total number of hits that the ISI Web of Knowledge will display at one time.

Drosophila 10,000 hits
BRAND AH, PERRIMON N - 4,088 citations

HIV 10,000 hits
Heil F, Hemmi H, Hochrein H, et al.- 1,169 citations

Evolution 10,000 hits
Bartel DP - 564 citations

Cancer 10,000 hits
Thiery JP, Acloque H, Huang RYJ, et al. - 131 citations

Ecology 8,930 hits
EMLEN ST, ORING LW - 2,782 citations

Population genetics 2,366 hits
Kumar S, Tamura K, Nei M - 7,437 citations

E.coli 1,484 hits
Datsenko KA, Wanner BL - 3,079 citations

C.elegans 1,125 hits
Bartel DP - 3,744 citations

Darwin 771 hits
BECKER PJ, COPPENS P - 1,243 citations